Someone asked me last week for a link to "the free Turnitin AI checker". I went looking for the signup page so I could send it. There is no signup page. The vendor FAQ says why in one line, and the line is more specific than "it costs money".
That answer is boring. What is not boring is the shape of the gate, because there are two of them stacked, and almost every explanation online only describes the first one.
Gate one: the feature is sold separately from the product
The line in the vendor's own detection FAQ is short:
"AI writing detection is only available to customers that license Turnitin Originality."
That sentence is doing more work than it looks like. It is not saying "institutions pay for Turnitin". It is saying the AI detector rides on one specific SKU. The same page then spells out who is on the wrong side of the line:
"If you are a Turnitin Similarity, Turnitin Feedback Studio (TFS) or Originality Check customer, please speak to your Turnitin account manager regarding access to AI writing detection."
So "my university uses Turnitin" and "my university can see an AI score" are two different statements. A school can be a paying customer of three named products and still not have the detector. The one alternative route that page names is also an upsell:
"iThenticate 2.0 customers can get access to this feature if they license AI writing capabilities as an add-on."
This is the part that gets flattened in every summary I have read. People write "it's institutional" and stop. The useful version is: it is institutional and optional within institutional, so the honest answer to "will my submission get an AI score" is "ask, because the product tier decides it and you can't see the tier."
Gate two: even with the license, the score has a role check
This is the one that surprises people. At an institution that does license it, students still cannot read the number. Two separate sentences on the same page say so:
"Please note, only instructors and administrators are able to see the indicator."
"The AI writing detection indicator and report are not visible to students."
If you think of it as authorization rather than pricing, it snaps into place immediately: the score is an attribute on the submission with read access scoped to two roles, and the submitter is not one of them. You submit the document, the classifier runs, the value gets written, and your view does not render it.
There is exactly one documented path for that value to reach you, and it is a human one:
"However, with the PDF download feature, instructors can download and share the AI report with students."
Which means: an instructor exports it and sends it to you, or you never see it. Whether that happens is a policy question at your institution, not a feature you can enable. Do not skip that half when you repeat the "students can't see it" line — plenty of students have been shown their report, and they were shown it this way.
Why the "free Turnitin checker" sites cannot answer your question
They are not running that model. They are their own classifiers, trained on their own data, sitting on a domain that bought the brand name as a search term.
I am not going to tell you which one is more accurate, because I have not measured that and neither has anyone quoting a number at you. The point is narrower and it holds regardless of accuracy: two different classifiers produce two different numbers on the same text, so a number from one is not a prediction of the other. If your actual question is "what will my institution's report say", a third-party score cannot answer it — not because it is bad, but because it is a different measurement.
There is one more thing worth knowing if you ever do get handed a report, and it is in the same FAQ:
"The Similarity score and the AI writing detection percentage are completely independent and do not influence each other."
Two numbers, two pipelines. A high similarity score does not push the AI number up and a low AI number says nothing about your citations.
What you can actually do
Not much on the detection side, and I would rather say that plainly than pad it out:
- Ask your instructor which products the institution licenses. That is the only way to find out whether an AI indicator exists on your submissions at all, and it is a reasonable question to ask.
- If you are told your paper was flagged, ask for the PDF. The export exists and it is the documented way for you to receive it. "I'd like to see the report" is a normal request, not an accusation.
- Stop comparing third-party percentages to each other. They are different systems. Watching one go from 40% to 30% tells you about that tool.
- Keep your drafts, notes and version history. That is the evidence that is actually yours, and unlike a score, nobody has to grant you access to it.
If a report does land in your hands and you want to work on the passages it marked, HumanPen takes the exported Turnitin or iThenticate report, matches the flagged passages back to your document, and rewrites only the ones you confirm — the rest of the file is not touched, and it is designed to preserve your citations, tables and formatting. Eligible passages can be re-run at no charge. It is at humanpen.net/humanize, and I work on it, so weigh that accordingly.
Every quote above is from Turnitin's AI writing detection capabilities FAQ, read in a browser on 15 September 2026.
Top comments (0)